The present invention relates to the enzyme aspartate decarboxylase, and in particular the use of its crystal structure for drug discovery.
Pantothenic acid (vitamin B5) is found in coenzyme A (CoA) and the acyl carrier protein (ACP), both of which are involved in fatty acid metabolism.
Pantothenic acid can be synthesised by plants and microorganisms but animals are apparently unable to make the vitamin, and require it in their diet. However, all organisms are able to convert pantothenic acid to its metabolically active form, coenzyme A.
The pathway for the synthesis of pantothenic acid in bacteria is shown in FIG. 1. It provides a potential target for the treatment of infectious disease, since inhibitors of the pathway should be damaging to microorganisms but not to human or animal subjects infected by microorganisms.
Of specific interest is aspartate decarboxylase (L-aspartate-xcex1-decarboxylase (EC 4.1.1.1)). This enzyme catalyses the decarboxylation of L-aspartate to xcex2-alanine, which then goes on to form pantothenate in a condensation reaction with D-pantoate. Inhibitors (whether competitive, non-competitive, uncompetitive or irreversible) of aspartate decarboxylase (ADC) would be of significant technical and commercial interest.
ADC was first isolated from Escherichia coli by Williamson et al. (J. Biol. Chem., 254, (1979), 8074-8082), who found indications that the protein was present in different processed states. The unprocessed enzyme is referred to as the n-chain and has 126 residues. Processing (see FIG. 2) splits the n-chain at the Gly24-Ser25 peptide bond into a larger C-terminal chain and a smaller N-terminal chain. A pyruvol group (for convenience termed Pv125) is generated from the serine residue (Ser25) at the end of the C-chain, and a carboxylate group is formed at the end of the glycine residue (Gly24) of the smaller N-terminal chain. Williamson et al. found that only a proportion of the enzyme chains were processed in this way.
Purification to homogeneity of overexpressed, recombinant ADC was achieved by Ramjee et al. (J. Biochem., 323, (1997), 661-669). The purified enzyme was found to be a tetramer which, after processing, contained three processed chains and one chain which was not fully processed.
Albert et al. (Nature Structural Biology, 5, (1998), 289-293) used X-ray crystallography to determine the structure of ADC to 2.2 xc3x85 resolution. They showed that the enzyme studied by Ramjee et al. has pseudo-fourfold rotational symmetry, each of the four tetramer subunits (each subunit or corresponding to a n-chain labelled A, B, C or D) having a six-stranded xcex2-barrel capped by small xcex1-helices at each end. The binding cavities for aspartate decarboxylation are located between adjacent subunits. Three of the binding cavities have catalytic pyruvol groups resulting from respective processed n-chains. The other binding cavity has an ester which appears to be an intermediate in the processing reaction. The evidence points to an autocatalytic self-processing mechanism which did not lead to full processing of all the n-chains. The coordinates of the crystal structure determined by Albert et al. are available from the Protein Data Bank (Berman et al., Nucleic Acids Research, 28, (2000), 235-242) under access code lAW8.
Albert et al. proposed a model of L-asparate binding, but did not suggest a mechanism by which ADC accomplishes aspartate decarboxylation. Until now very little was known about the enzyme""s role in catalysis. This has impeded the development of ADC inhibitors via structure-based drug design methodologies. Knowledge of the mechanism would significantly assist the rational design of novel therapeutics based on ADC inhibitors.
Specific residues are denoted herein by their conventional acronyms (e.g. Gly for glycine), and numbers corresponding to their position in the unprocessed n-chain counting from the N-terminal of the n-chain (e.g. Gly24). Moreover, because each binding cavity is formed from the residues of two n-chains, each residue is further denoted by a letter corresponding to the respective one of the n-chains (e.g. Gly24A or Lys9D). Below, we have used D and A to denote the two n-chains of a binding cavity, but in a tetramer with four equivalent binding cavities and subunits labelled A, B, C and D one could equally use A and B, B and C, or C and D instead.
In the following by xe2x80x9cbinding sitexe2x80x9d we mean a site, such as an atom or functional group of an amino acid residue, in the ADC binding cavity which may bind to an agent compound such as a candidate inhibitor. Depending on the particular molecule in the cavity, sites may exhibit attractive or repulsive binding interactions, brought about by charge, steric considerations and the like.
By xe2x80x9cfittingxe2x80x9d, is meant determining by automatic, or semi-automatic means, interactions between one or more atoms of an agent molecule and one or more atoms or binding sites of the ADC, and determining the extent to which such interactions are stable. Various computer-based methods for fitting are described further herein.
By xe2x80x9cfully processedxe2x80x9d ADC we mean a composition comprising an amount of ADC in which pyruvoyl groups are generated from at least 90%, preferably at least 95%, and more preferably at least 99% of the ADC Ser25 residues.
By xe2x80x9croot mean square deviationxe2x80x9d we mean the square root of the arithmetic mean of the squares of the deviations from the mean.
By a xe2x80x9ccomputer systemxe2x80x9d we mean the hardware means, software means and data storage means used to analyse atomic coordinate data. The minimum hardware means of the computer-based systems of the present invention comprises a central processing unit (CPU), input means, output means and data storage means. Desirably a monitor is provided to visualise structure data. The data storage means may be RAM or means for accessing computer readable media of the invention. Examples of such systems are microcomputer workstations available from Silicon Graphics Incorporated and Sun Microsystems running Unix based, Windows NT or IBM OS/2 operating systems.
By xe2x80x9ccomputer readable mediaxe2x80x9d we mean any media which can be read and accessed directly by a computer e.g. so that the media is suitable for use in the above-mentioned computer system. The media include, but are not limited to: magnetic storage media such as floppy discs, hard disc storage medium and magnetic tape; optical storage media such as optical discs or CD-ROM; electrical storage media such as RAM and ROM; and hybrids of these categories such as magnetic/optical storage media.
The present invention is at least partly based on overcoming several technical hurdles: we have (i) produced fully processed crystals of ADC of suitable quality for performing X-ray diffraction analyses, (ii) formed ADC-ligand complexes by soaking the crystals in appropriate soaking solutions, (iii) collected X-ray diffraction data from the ADC-ligand complexes, (iv) determined the three-dimensional structures of the complexes, (v) identified regions of ADC which undergo conformational changes upon ligand binding and decarboxylation, and (vi) determined the likely mechanism by which ADC accomplishes aspartate decarboxylation.
In general aspects, the present invention is concerned with identifying or obtaining agent compounds (especially inhibitors of ADC) for modulating ADC activity, and in preferred embodiments identifying or obtaining actual agent compounds/inhibitors. Crystal structure information presented herein as useful in designing potential inhibitors and modelling them or their potential interaction with the ADC binding cavity. Potential inhibitors may be brought into contact with ADC to test for ability to interact with the ADC binding cavity. Actual inhibitors may be identified from among potential inhibitors synthesized following design and model work performed in silico. An inhibitor identified using the present invention may be formulated into a composition, for instance a composition comprising a pharmaceutically acceptable excipient, and may be used in the manufacture of a medicament for use in a method of treatment. These and other aspects and embodiments of the present invention are discussed below.
A first aspect of the invention provides a crystal of fully processed ADC having a hexagonal space group P6122, and unit cell dimensions of a=71.1 xc3x85, and c=215.8 xc3x85, or more generally a=71.1xc2x10.2 xc3x85, and c=215.8xc2x10.2 xc3x85.
Alternatively or additionally, the crystal has the three dimensional atomic coordinates of Table 1. An advantageous feature of the structural data according to Table 1 are that they have a high resolution of about 1.55 xc3x85.
The coordinates of Table 1 provide a measure of atomic location in Angstroms, to a first decimal place. The coordinates are a relative set of positions that define a shape in three dimensions. It is possible that an entirely different set of coordinates having a different origin and/or axes could define a similar or identical shape. Furthermore, varying the relative atomic positions of the atoms of the structure so that the root mean square deviation of the conserved residue backbone atoms (i.e. the nitrogen-carbon-carbon backbone atoms of the protein amino acid residues) is less than 1.5 xc3x85 (preferably less than 1.0 xc3x85 and more preferably less than 0.5 xc3x85) when superimposed on the coordinates provided in Table 1 for the conserved residue backbone atoms, will generally result in a structure which is substantially the same as the structure of Table 1 in terms of both its structural characteristics and potency for structure-based drug design of ADC inhibitors. Likewise changing the number and/or positions of the water molecules of Table 1 will not generally affect the potency of the structure for structure-based drug design of ADC inhibitors. Thus for the purposes described herein as being aspects of the present invention, it is within the scope of the invention if: the Table 1 coordinates are transposed to a different origin and/or axes; the relative atomic positions of the atoms of the structure are varied so that the root mean square deviation of conserved residue backbone atoms is less than 1.5 xc3x85 (preferably less than 1.0 xc3x85 and more preferably less than 0.5 xc3x85) when superimposed on the coordinates provided in Table 1 for the conserved residue backbone atoms; and/or the number and/or positions of water molecules is varied. Reference herein to the coordinates of Table 1 thus includes the coordinates in which one or more individual values of the Table are varied in this way.
Also, modifications in the ADC crystal structure due to e.g. mutations, additions, substitutions, and/or deletions of amino acid residues (including the deletion of one or more tetramer subunits) could account for variations in the ADC atomic coordinates. However, atomic coordinate data of ADC modified so that a ligand that bound to one or more binding sites of ADC would also be expected to bind to the corresponding binding sites of the modified ADC are, for the purposes described herein as being aspects of the present invention, also within the scope of the invention. Reference herein to the coordinates of Table 1 thus includes the coordinates modified in this way. Preferably, the modified coordinate data define at least one ADC binding cavity.
We have been able to produce and isolate for the first time fully-processed ADC, in which the binding cavities of substantially all the ADC molecules are identical and each binding cavity has a catalytic pyruvol group. This has been made possible by the identification of conditions which allow the processing reaction to proceed to completion.
A second aspect of the invention provides a method of fully processing ADC comprising the step of forming a solution of ADC, the solution having a pH in the range 6.5-8.5 (preferably 7.0-8.0) and an ADC concentration in the range 1-50 mg/ml (preferably 4-20 mg/ml).
The method may further comprise the step of crystallising the dissolved ADC to form a crystal of fully processed ADC.
In a third aspect, the invention provides a method of testing a candidate agent compound (such as a candidate inhibitor of ADC) for ability to modulate ADC activity comprising the step of contacting the candidate agent compound with fully processed ADC (produced e.g. according to the method of the second aspect) to determine the ability of the candidate agent compound to interact with ADC.
Preferably, the candidate agent compound is contacted with ADC in the presence of L-aspartate, and typically a buffer.
By using fully processed ADC for forming ADC-ligand complexes more candidate agent compound molecules per molecule of ADC are exposed to fully processed binding cavities, thereby increasing the sensitivity of e.g. chemical assays based on such complexes.
In fourth aspect, the invention provides a method of analysing a fully processed ADC-ligand complex comprising the step of employing (i) X-ray crystallographic diffraction data from the fully processed ADC-ligand complex and (ii) a three-dimensional structure of fully processed ADC, to generate a difference Fourier electron density map of the complex, the three-dimensional structure being defined by atomic coordinate data according to Table 1.
Electron density maps can be calculated using programs such as those from the CCP4 computing package (Collaborative Computational Project 4. The-CCP4 Suite: Programs for Protein Crystallography, Acta Crystallographica, D50, (1994), 760-763.). For map visualisation and model building programs such as O (Jones et al., Acta Crystallograhy, A47, (1991), 110-119) can be used.
In a fifth aspect, the invention provides a method of identifying an agent compound (such as an inhibitor of ADC) which modulates ADC activity comprising the steps of:
a) providing a candidate agent compound;
b) forming a complex of fully processed ADC (produced e.g. according to the method of the second aspect) and the candidate agent compound; and
c) analysing said complex by X-ray crystallography (e.g. according to the method of the fourth aspect) or by NMR spectroscopy to determine the ability of said candidate agent compound to interact with ADC. Detailed structural information can then be obtained about the binding of the agent compound to ADC, and in the light of this information adjustments can be made to the structure or functionality of the agent compound, e.g. to improve binding to the binding cavity. Steps b) and c) may be repeated and re-repeated as necessary. For X-ray crystallographic analysis, the complex may be formed by crystal soaking or co-crystallisation.
Therefore, compared to partially processed ADC, X-ray crystallographic data from the binding cavities of fully processed ADC-ligand complexes can be interpreted more easily because all the binding cavities are identical. That is, the data are not complicated by reflections from binding sites containing esters instead of pyruvol groups. Likewise the interpretation of NMR spectra is simplified.
In a sixth aspect, the present invention provides a method of identifying an agent compound (such as an inhibitor of ADC) which modulates ADC activity, comprising the steps of:
a) providing a model of a binding cavity of ADC, said model including at least one (and preferably both) of binding site nos. 1 and 9 defined by Table 2;
b) providing the structure of a candidate agent compound;
c) fitting the candidate agent compound to said binding cavity, including determining the interactions between the candidate agent compound and at least one (and preferably both) of binding site nos. 1 and 9; and
d) selecting the fitted candidate agent compound.
Without wishing to be held to any particular theory, we believe that, in the appropriate context (e.g. in the complexes described below in the xe2x80x9cDetailed Description of the Inventionxe2x80x9d), one or more of the binding sites of Table 2 provides the corresponding binding interaction of Table 2 to an agent compound. However, the binding interactions of Table 2 are not intended to be exhaustive, and it is within the scope of this aspect of the invention that any of the binding sites may exhibit an interaction which is not listed in Table 2.
Varying the relative positions of the binding sites of Table 2 by relatively small amounts generally results in arrangements of binding sites which are substantially identical to the arrangement of Table 2 in terms of expected interactions with the agent compound. Consequently, the scope of this aspect of the invention includes a binding cavity in which the root mean square deviation of the conserved residue backbone atoms of the residues of column 2 of Table 2 is less than 1.5 xc3x85 (preferably less than 1.0 xc3x85 and more preferably less than 0.5 xc3x85) when superimposed on the coordinates provided in Table 1 for the conserved residue backbone atoms of the residues of column 2 of Table 2.
The smaller N-terminal xcex2-chain has a tail (hereafter called Tail24A) formed when the n-chain cleaves at the Gly24-Ser25 peptide bond and consisting of the four residues His2A, Tyr22A, Glu23A, and Gly24A (as discussed above, Gly24A having a carboxylate end group). We have found that Tail24A shifts between an xe2x80x9copenxe2x80x9d and a xe2x80x9cclosedxe2x80x9d position via a xe2x80x9chalf-closedxe2x80x9d position (which we call the O-state, C-state and H-state respectively) during aspartate decarboxylation. In the C-state Tail24A obstructs the binding cavity, while the O-state allows access thereto. These states are characterised by increased disorder in the measured position of Tail24A as it shifts from the C-state to the O-state.
Binding site no. 1 is associated with the hydrophobic phenyl ring of Tyr22A which in turn belongs to Tail24A. Hence binding site no. 1 is closely involved with the C-, H- and O-states of Tail24.
The NH3+ group (binding site no. 9) of the Lys9D side chain is a potential hydrogen bond donor when Tail24A is in the O- and H-states. However, we have found that in the C-state the Gly24A carboxylate end group forms a salt bridge or hydrogen bond with the NH3+ group of the Lys9D side chain. This prevents the NH3+ group from being a potential hydrogen bond donor to the agent compound in the C-state.
The modelling may include generating the cavity (and optionally the agent compound) on a computer screen for visual inspection.
In practice, it is desirable to model a sufficient number of atoms of the ADC as defined by the coordinates of Table 1. Thus, in this aspect of the invention, there will preferably be provided the coordinates of at least 5, preferably at least 10, more preferably at least 50 and even more preferably at least 100 atoms of the ADC structure.
Preferred candidate agent compounds bind with at least two, three, four, five, six or seven of the binding sites defined by Table 2. In general, the agent compound binds better as the strength and number of binding interactions increases. The candidate agent compound may have a molecular weight of up to about 600.
Binding interactions may be mediated by e.g. water or other solvent molecules.
Candidate inhibitors identified according to the method are characterised by their suitability for binding to a particular binding site or sites. The binding cavity can therefore be regarded as a type of binding site framework or negative template with which the candidate inhibitors correlate in the manner described above.
More specifically, a potential modulator of ADC activity can be examined through the use of computer modelling using a docking program such as GRAM, DOCK, or AUTODOCK (see Walters et al., Drug Discovery Today, Vol.3, No.4, (1998), 160-178, and Dunbrack et al., Folding and Design, 2, (1997), 27-42) to identify candidate inhibitors of ADC. This procedure can include computer fitting of candidate inhibitors to ADC to ascertain how well the shape and the chemical structure of the candidate inhibitor will bind to the enzyme.
Computer programs can be employed to estimate the interactions between the ADC and the agent compound. The more specificity in the design of a candidate drug, the more likely it is that the drug will not interact with other proteins as well. This will tend to minimise side-effects due to unwanted interactions with other proteins.
Alternatively, step b) of the method may involve selecting the candidate agent compound by computationally screening a database of compounds for interaction with the binding cavity. For example, the model resulting from step a) may be used to interrogate the compound database, a candidate inhibitor being a compound that has a good match to the features of the model. In effect, the model is a type of virtual pharmacophore.
If one or more additional ADC binding cavities are characterised and a plurality of respective compounds are designed or selected, the candidate inhibitor may be formed by linking the respective compounds into a larger compound which maintains the relative positions and orientations of the respective compounds at the binding cavities. The larger compound may be formed as a real molecule or by computer modelling.
Having determined possible binding partners, these can then be obtained or synthesised and screened for activity. Consequently, the method preferably comprises the further step of:
e) contacting the candidate agent compound with ADC to determine the ability of the candidate agent compound to interact with ADC.
Preferably, in step e) the candidate agent compound is contacted with ADC in the presence of L-aspartate, and typically a buffer.
Instead of, or in addition to, performing a chemical assay, the method may comprise the further steps of:
e) forming a complex of ADC and said candidate agent compound; and
f) analysing said complex by X-ray crystallography (e.g. according to the method of the fourth aspect) or by NMR spectroscopy to determine the ability of said candidate agent compound to interact with ADC. Detailed structural information can then be obtained about the binding of the candidate agent compound to ADC, and in the light of this information adjustments can be made to the structure or functionality of the candidate agent compound, e.g. to improve binding to the binding cavity. Steps e) and f) may then be repeated and re-repeated as necessary. For X-ray crystallographic analysis, the complex may be formed by crystal soaking or co-crystallisation.
In another aspect, the invention includes a compound which is identified as an agent compound (such as an inhibitor of ADC) for modulating ADC activity by the method of one the previous aspects.
Following identification of an agent compound it may be manufactured and/or used in preparation, i.e. manufacture or formulation, of a composition such as a medicament, pharmaceutical composition or drug. These may be administered to individuals.
Thus, the present invention extends in various aspects not only to an agent compound as provided by the invention, but also a pharmaceutical composition, medicament, drug or other composition comprising such an agent compound e.g. for treatment (which may include preventative treatment) of a disease such as a microbial infection; a method comprising administration of such a composition to a patient, e.g. for treatment of a disease such as a microbial infection; use of such an agent compound in the manufacture of a composition for administration, e.g. for treatment of a disease such as a microbial infection; and a method of making a pharmaceutical composition comprising admixing such an agent compound with a pharmaceutically acceptable excipient, vehicle or carrier, and optionally other ingredients.
In another aspect, the present invention provides a system, particularly a computer system, intended to generate structures and/or perform rational drug design for ADC or ADC ligand complexes, the system containing either (a) atomic coordinate data according to Table 1, said data defining the three-dimensional structure of fully-processed ADC, or (b) structure factor data for fully-processed ADC, said structure factor data being derivable from the atomic coordinate data of Table 1.
In a further aspect, the present invention provides computer readable media with either (a) atomic coordinate data according to Table 1 recorded thereon, said data defining the three-dimensional structure of fully-processed ADC, or (b) structure factor data for fully-processed ADC recorded thereon, the structure factor data being derivable from the atomic coordinate data of Table 1.
By providing such computer readable media, the atomic coordinate data can be routinely accessed to model fully-processed ADC. For example, RASMOL (Sayle et al., Trends in Biochemical Sciences, Vol. 20, (1995), 374) is a publicly available computer software package which allows access and analysis of atomic coordinate data for structure determination and/or rational drug design.
On the other hand, structure factor data, which are derivable from atomic coordinate data (see e.g. Blundell et al., Protein Crystallography, Academic Press, New York, London and San Francisco, (1976)), are particularly useful for calculating e.g. difference Fourier electron density maps.